craig.cahillane@LIGO.ORG - posted 04:04, Wednesday 06 February 2019 - last comment - 08:56, Wednesday 06 February 2019(46817)
Intensity Noise Projections into DARM
Since the arm power measurement revealed that there might be an 8% discrepancy between the two arm powers, with the Y arm having more power, we have become concerned that Input RIN coupling to DARM may be worse than anticipated. Last evening I took some broadband intensity noise injections and an intensity noise swept sine when I couldn't get coherence at low frequency. Today I projected these noises into DARM. (attachment one) Although low frequency intensity noise is higher than it should be if our arms were balanced, we are not limited by input intensity noise in DARM except at 3 kHz and upward. Attachment two is the DARM/Input RIN TF from the broadband injection. Attachments three, four, and five are the DARM/Input RIN TF, Arm RIN/Input RIN TF, and DARM/Arm RIN TF from the swept sines. These are reposted from alog 46781 for convenience.
Images attached to this report
Comments related to this report
Here's a comparison with the simulation results described in 46759.
- DARM / RIN: at low frequency the coupling goes like f^-3 as expected from simulation (f^-1 due to the double cavity pole and f^2 for the test mass response to radiation pressure). The amplitude is about correct too. At 20 Hz the simulation value is 2.3e-12 m/(W/W) while the measurement is 3.7e-12 m/(W/W). The ratio is 8/5, which correspond precisely to the ratio of measured arm power imbalanced and simulated power imbalance.
- In other words, the measured low frequency transfer function matches the expectation from radiation pressure when the power imbalance in the arm is 8%
- ARM_RIN / INPUT_RIN should give a measurement of the double cavity pole. The value 0.56 Hz is very close to the expected 0.6 Hz.
- DARM / ARM_RIN gives a measurement of the power imbalance. The measured value gives dP = 13.1 kW. The expected transfer function is
where the mass is 39.6 kg and the power imbalance value is extracted from the measured TF at 20 Hz
Images attached to this comment